The Brilliant World of FX (Episode 3): A tour around the world with Setser and Pettis

Welcome to PodCept and the brilliant world of FX, the podcast where Deutsche Bank's research team debates the global macro and currency outlook. I'm George Saravelos, head of FX research here at Deutsche Bank, and today I'm delighted to be bringing two famous guests on board, Brad Setzer, who is Senior Fellow at the Council of Foreign Relations in the US, and he's the world's supreme expert on global capital flows and balance of payments. And also joining me is Michael Pettis, Professor of Finance at Peking University in China.

He is a world authority on China. The recording you will hear is from a panel discussion at Deutsche Bank's macro conference in New York a few days ago. The sound quality may be a little different, but the content hopefully compensates.

And with that, over to the conference recording. Okay, thank you very much. My name is George Saravelos.

I'm the head of the FX research team here at Deutsche Bank. I'm tremendously excited to welcome Brad Setzer and Michael Pettis. Let's get right into it, and let's start with China.

Huge focus on AI, NASDAQ's making record highs, but Chinese equities are also making even bigger record highs. So Michael, if you could just contextualize what is going on in China macro over the last six months, and then this anti-involution, which is the new word that keeps getting repeated. Just talk to us a bit about that and what's going on in the ground.

As far as the equity markets go, at the beginning of last year, I told a bunch of my clients that I thought now was a good time to buy Chinese equity. The first six months, as you can imagine, those of you who follow on a day-to-day basis will know that the first six months, it was a really, really tough call. And then luckily, things started to go up.

But does that mean that I was expecting fundamentals to improve or growth to pick up or profits to go up? No, not at all. The stock markets in China have nothing to do with any of that.

What really drives them is partly domestic liquidity and partly the credibility of government signaling. And so in the beginning of last year, it seemed pretty clear that the government was quite desperate to get the markets to go up again because they felt it was necessary to boost confidence and they started unleashing enough measures that made several people convinced that they were serious and they were. So we have a national team.

Every time the stocks go down by some amount, we don't know what the amount is. The national team comes in and buys. So you have a very asymmetrical payoff, right?

You have lots of upside and you assume your downside is pretty limited. And under those conditions, you should probably buy. I think it's probably still good for at least a few more months, but we'll see.

It's really a credibility issue. But in terms of the economy, the economy is not getting better. It's in a pretty tight position, which we can discuss as part of the evolution problem.

And business profits, as you know, are either down or growing more slowly than nominal GDP. So it's really not a profit-driven story in the stock market. It's really a government signaling-driven story.

The thing about involution, nezhuan, which is like the, since May has become the really big word, is just a new name for a very old problem. And the problem, you can call it excess capacity. We're not allowed to say that anymore.

So it's involution. And it's a problem that's really existed for 15 years. So the way I think about it is the decision to invest in China is not based on any perceived investment needs.

It's really quite a simple formula, right? You set the GDP growth target, which is set politically. By the way, we can tell you, some people can tell you what the GDP growth rate for China will be for the next 10 years.

Because officially, from 2020 to 2035, GDP had to double. And you actually hear people say, senior policy makers say, that we will grow by whatever it is, 4.5% a year for the next 10 years because we have to double GDP. That's sort of a scary thought, but that's how many people think.

But once you set the GDP growth target, GDP growth is equal basically to consumption growth plus investment growth, with net exports being the residual. And if you can't get consumption growth to rise, and they can't, and I'm sure we'll discuss that later, then you cannot let investment growth fall, right? So the GDP growth target is also an investment growth target, regardless of the actual investment needs of the economy.

So you'll remember during the great financial crisis, China's current account surplus collapsed. It was a little bit above 10%, and it dropped to 3%. In order to counter that, again, many of you may remember the famous $4 trillion fiscal expansion, most of that went into a huge surge in infrastructure investment.

And the point that I want to make, and I'm going to make it several times, is that nobody in Beijing in 2010 said, my God, we don't have enough infrastructure. Let's build infrastructure. And isn't it lucky it occurred at the same time as the collapse in the current account surplus?

It was a direct consequence of the collapse in the current account surplus. So if you look at the data, infrastructure investment was extremely high, and we started vastly overbuilding. That came to an end around the middle of the decade.

There were a number of problems in the middle of the decade, and the response was to engineer a surge in the property sector. And so they cut mortgage rates, they cut the minimum deposits, they cut restrictions on owning apartments, and we got the latest and what turned out to be the last stage of the property bubble. So again, the decision to increase property investment was not based on any understanding that there weren't enough apartment buildings in China.

It was really a reaction to the slowdown in the underlying economy. That came to an end, of course, in 2021-22, and property investment dropped very, very sharply. Property was roughly a little bit more than a third of total investment, dropped very sharply, and because the GDP growth target could not allow a drop in total investment, we saw an almost dollar-for-dollar increase in manufacturing investment.

And again, nobody in Beijing said, there is so much global demand for manufacturing and not enough capacity, let's double or triple capacity. That was not the decision. The decision was you couldn't allow investment growth to drop.

Investment growth in property was dropping sharply, so it had to go up somewhere else. There was too much infrastructure, let's do manufacturing. The thing is, at the time, there were a number of areas that Beijing had decided were the sort of new industrial future, solar panels, electric vehicles, batteries, and local governments in their desire to really cultivate favor concentrated most of the increase in manufacturing investment in those sectors.

So it always astonishes me because by the end of 2023, solar panels, there was already excess capacity in solar panels, and it doubled in 2024. And in the first half of 2025, it went up again by something like a third, by which time solar panel factories were well below 40% capacity. Everybody was losing money.

Solar panels were being sold, and I learned in economics that this can't happen, but it can. They were being sold below variable cost because you had to keep producing and you couldn't continue financing rising inventories because clearly the banks weren't that interested in doing it. So you have to sell no matter what the cost, and so you got this really vicious price competition.

And that became such a problem that around May and June, we started seeing a movement towards resolving involution. And I think they have resolved it. If you look at the July and August numbers, industrial production is growing, but much lower than expected.

Exports are growing, but much lower than expected. Deflation has sort of disappeared. I don't think it's disappeared, but it's temporarily gone.

We had a small price increase in July and flat prices in August, all of which suggests that they were able to bring involution under control. But remember, involution was created to resolve the problem of declining investment in the property sector, and that's not recovering. So if you reduce investment in involution in the involuted areas and you still want to meet the GDP growth target, we're going to need excess capacity to shift.

So we've already gone from infrastructure to property to involuted manufacturing. My guess is it'll go to non-involuted manufacturing and additional infrastructure. So I think the optimist would say, well, maybe it will go into consumption, and we can get to that.

But I think all your points nicely feed into the whole global imbalances question. Brad, you've spent a lot of time and effort talking about China being the big source of excess supply, but contextualize that in the global framework. And is really China, I mean, the Trump administration is obviously very focused on China.

Is China the epicenter of the whole global imbalance problem from your perspective? Just give us a sense of magnitude. Well, I actually wish the Trump administration were a bit more focused on China.

Well, they were. I mean, they kind of were, then they weren't. And now we actually have higher tariffs on Brazil and India than on China.

So I'm a bit confused about Trump's trade policy and how it relates to this debate about imbalances. I have no doubt, though, that China is the center of imbalances. Survey the world 10 years ago, you might say roughly that Europe has a pretty big surplus.

Much of East Asia outside of China has a pretty big surplus. Depends on where the price of oil is. But when oil prices are high, the oil guys had a pretty high surplus.

China was one of many. If you just look at goods trade and there is a stupid statistical issue that I won't bore you with. Just looking at goods trade right now, global imbalances on the surplus side are essentially found in China, Korea and Taiwan.

Japan's trades balance Europe's trade, if you take out weird pharmaceuticals, but take out Irish pharmaceutical surplus to the US, Europe's goods trade is pretty close to balance. Canada, Mexico, pretty close to balance. So you really see an incredible concentration of the surplus on in East Asia and in China in particular.

The goods surplus has gone from 400 billion to 1.2 trillion over the past five years. Past couple of years, Chinese export growth, 24, was sort of 12, 13 percent in volume terms. This year, it is tracking to be above eight, eight seems like it's coming down as decelerated.

But in a world where global trade is growing at three percent, China is massively outperforming global trade and taking export share from pretty much everyone, but particularly Europe. So I think increasingly you can say that China's solution to the property bust was on one hand to direct money into manufacturers, solar panels, build the new EV industry when you had an ICE, a conventional car industry, but also to pivot it hard back to exports with between one third and one half, depending on the quarter of Chinese growth coming from net exports. So on one hand, we have kind of a world that has never been more protectionist than it is today.

In any measure of protectionist action, we will be at a peak. And on another level, the world's second biggest economy has not been this dependent on exports for growth, I think, ever, because back in the first China shock period, the domestic drivers of growth were also quite strong. Now we have a China where the domestic drivers of growth are pretty weak.

And so almost an enormous component of the impulse that has propelled China's economy in the last three years has come from exports. So I'll stick with you, Brad, and ask a market question, which is on the exchange rate, because you have a situation where trillions are literally being generated in dollar export receipts or euro export receipts, yet the currency is still very weak. And when you look at central bank intervention, actually reserves are going sideways.

So can you help us square that circle? And the currency should be massively appreciating, theoretically, given such a massive surplus. It is an interesting question.

I mean, actually, if you look at the detailed analysis of the Chinese central bank reserves, they're kind of going down, they're modestly down, they're down $10 billion a month on the PBOC balance sheet, presumably that's some reallocation to gold. So it's even a bigger puzzle. It's not just that reserves are stable, they've actually been trending down.

So what has generated an equilibrium where China's currency is actually extremely low? It's depreciated by close 15 percent in real effective terms the past few years, contributing to this export boom, despite an enormous increase in the economy's net dollar generation? I think the answer is basically threefold.

One is Chinese rates are low, well below U.S. rates. And so there's an incentive, absent appreciation, to hold dollars, to hoard dollars. You see that in big increases in onshore dollar deposits and presumably offshore dollars throughout Asia.

Every indicator that I've looked to about Hong Kong banking system liquidity, either in dollars or Hong Kong dollars, suggests that it is enormously liquid, almost as liquid as it is enormously liquid, almost certainly because of offshore Chinese money. And some of that is just exporters not wanting to convert. Second factor is the direction of FDI flows has completely flipped.

Five years ago, foreign investors wanted to put money into China and foreign investors who had money in China wanted to keep the money in China to pick up on the higher CNY rates than you could get at that time on dollar rates. After 2022, foreign investment, direct investment in China tanks and China liberalizes. It's easy to get money out if you're setting up a factory abroad.

She encourages this big source of flows. Then the third is, guess who, Michael talks about the national team. Whenever right now the CNY is at the fix, not the top edge of the fix, at the fix.

And if you look at over the past 12 months, the CNY has never traded strong in this band. And what you see is at the fix, the state banks apparently are just consistently coming in. You see big increases in FX settlement, which is a number that includes state banks and the PBOC.

You see big buildups in the state banks, foreign assets. So in a sense, the answer is there's a big increase in Chinese willingness to hold dollars in foreign assets because of the rate differential. That differential is no longer sufficient.

And when it is no longer sufficient, the PBOC sends a signal to the state banks, and then the state banks seem to do all the dirty work. It's an interesting world because we speak about deglobalization, and yet the dollar remains massively dominant in that part of the world in terms of the recycling. Well, the other interesting thing, and I'll stop here, is it is almost certain that the state banks have a much higher share of their assets in dollars than SAFE does.

So there's this funny little game where SAFE says it is de-dollarized, and it is down to 55% in dollar share of its reserves, yet more and more of the assets are on the state banks and in dollars. And an interesting dynamic is some of those assets are being deployed in the Belt and Road Initiative in Africa, and we see that all the time. So it's globalization, but in a different time.

Presumably, some of them are funded in basis trades in London. Who knows? So that was, I guess, the global element, but if we think ahead, I think there's a question mark on what happens to the currency.

But just the broader macro, where are we in two, three years' time? I think with China, there's always a discussion on what they should do versus what they will do, and the question is what they will do. So where will the growth rate be?

Will they bring the consumption share of GDP higher? Will the current account surplus stay as wide? We sort of know what they will do because every country that runs imbalances of nearly the size, and China has imbalances of historical records.

We've never seen such deep imbalances, but it's happened many times in the past. China didn't invent the high savings, high investment growth model. Many countries have followed it, and we always get rebalancing.

So the question is not, will China rebalance? The question is, how will it rebalance? And the historical precedents are pretty gloomy.

So you've got production way above consumption, and you need somehow to get them the right relationship between the two. If China were to raise its consumption share of GDP by 10 percentage points, it would still be the lowest consuming major economy in the world. We've got a long ways to go, but how do you close the gap?

Well, historically, we have two models, right? One I call the American model, because in the 1920s, the US had many of those problems. Very high savings, very high trade surplus, low domestic demand.

And how did it solve the problem? Well, a collapse in production and a smaller collapse in consumption, right? The household income share of GDP dropped by roughly half of the GEP, right?

So we saw a rebalancing, very brutal, very quick. But that is arithmetically one way you can rebalance. The second way that we've seen, I call for obvious reasons, the Japanese model.

And that is, so you've got production up here, consumption here. If you look at what happened in Japan after 1990, 1991, the consumption growth rate barely declined. It declined just a little bit, right?

And maybe on a per capita basis, it didn't decline. It's hard to say. And GDP growth dropped to basically zero.

So by definition, investment growth was negative. Not a very socially or politically disruptive adjustment, but it did take 30 years, 20 years, depending on how you think about it. And during that time, the Japanese share of global GDP went from 17%, which is where China is today, to about 9% 20 years later.

And I think today it's around 5% or 6%. So you could argue that the Japanese version of the adjustment is much less disruptive in the short term, but perhaps worse economically over the long term. And those are the only two models we have.

There is, in theory, a third model. So you've got production up here, consumption here. Why not get a surge in consumption?

And you rebalance these higher GDP growth rates. Theoretically possible. The question is...

That's what the Chinese want to do? Yes. Yes.

And they're arguing that they can do it because of the really heavy investment in technology, which will cause a productivity surge. There's two problems with that. One is that that's what the Soviets said in the 1960s.

And that's what the Japanese said in the 1980s. Obviously, it didn't work in either case. The second problem is that you can do the arithmetic.

I'm really surprised people haven't done the arithmetic because it's very simple. An Excel model will do it all in 10 minutes. Let's say I want China to rebalance by 10 percentage points in 10 years, which I think is pretty generous, right?

You can calculate the growth rate of consumption. If you want China to grow at 4.5% during that period, consumption has to grow at 6.4%. It doesn't seem very complicated.

So basically, we need consumption growth to rise by nearly 2 percentage points. Is that possible? It's happened in the past.

Certain really transformative technological innovations have managed to get productivity growth by 2%, in which case you need a couple of things. You need 100% of the increase in productivity to go to the household sector. And you also need, because this is a global imbalance, for the rest of the world not to participate in that productivity increase, right?

So it's pretty tough to count on that. The other way they can do it is, unlike many other countries, China has a huge government sector. Local governments own assets worth about 40-50% of GDP by some estimates.

So they could transfer income from local governments or transfer assets to the household sector. If they do, I calculated, it's something like 1.5 percentage points of GDP every year. You could, in theory, get consumption to grow at 6% or 7%, right?

Will they do it? There have been some proposals by prominent Chinese economists, very carefully worded, of course, but there have been some proposals. And the obvious objection is that that involves a sort of a transformation of political institutions, which may or may not be possible.

I don't know. But arithmetically, those are the three ways you rebalance. And only two of them we've actually seen in reality.

That was a really useful way of framing. And I'll pick up on the Japan framing, because Japan's growing its own transformation at the moment. Our inflation, we just got a new prime minister.

So let's leave China, still stay in Asia. I want to talk a bit about Taiwan and Japan. Brad, first of all, to say you wrote this FT piece on Taiwan.

Was it February of this year? With Josh Unger, who will be on a panel later today. And I think in that piece, you said, the risk is you see a massive rallying in Taiwan.

And unlike sell-side pieces, you actually wrote it before it happened, three months later. Now, since that, I mean, what you identify there is, again, current account recycling. It's a slightly different story to China, because the recycling is going to treasuries.

Same thing in Japan. But since that move, effectively, the recycling is continuing. And you're seeing Japan, Taiwan, they've not participated in this dollar rebalancing, which you've seen maybe in Europe.

So what's your view on that? Will that, you've said it's not a sustainable equilibrium, but it seems like we are going back to that equilibrium of just huge ongoing accumulation of dollar assets. Is similar story in Taiwan, Japan slightly different?

Well, you're nice, because I also wrote something in June that implied that the Taiwan dollar would likely continue to appreciate. And since then, it's stalled. And if anything, depreciated.

They are slightly different stories. Taiwan is the, Taiwan's imbalances make China's imbalances look like a little boy's game. Taiwan is running and has run current account surpluses that are probably averaging close to 15% of GDP.

And they are doing so without being a tax. No one is, you know, Apple does not book all of its profits in Taiwan. The pharmaceutical industry is not booking all of its profits in Taiwan.

This is a real chip production. Yeah, chip production, TSMC, da, da, da, da. The recycling has changed.

It used to be that the central bank would do a little backdoor intervention, not disclose it, do a swap with the lifers, and the lifers would buy foreign assets. And then the lifers also bought a bunch of foreign assets unhedged. That more or less came to an end during COVID.

And now the lifers are sitting on a portfolio of bonds before COVID that have relatively low interest rates, that are underwater, and that they can't sell. So the lifers are no longer central to this recycling story. So what has happened?

Well, US rates are way above Taiwan dollar rates. So there are periods of time when dollars accumulate in the domestic banking system. What else has happened?

Well, TSMC has been coerced into investing in the United States. And that is also generating a certain outflow. And then during certain periods of time, when there's a shift in expectations, and all it takes is a modest shift in expectations because the lifers are pretty much bust if they aren't hedged and there's a big move.

TSMC was very willing to hold dollars offshore so long as it thinks the Taiwan dollar is going to be stable, but it doesn't want to be caught if there's a move. So there's this potential for a big strengthening of the Taiwan dollar, but there's one big block. And what you saw in the second quarter is that when that pressure materializes, the central bank steps in.

Japan's a bit different. Japan doesn't actually want to trade surplus anymore. Current account surplus is all investment income, mostly FDI and mostly recycled by the big Japanese companies.

And coupons from treasuries. And coupons from the treasuries, but FDI is a bit bigger. Japan has not been a big buyer of global bonds for the past four years.

They actually sold in 22 when hedging costs went negative. Not been big, big, big buyers since. You don't actually historically need Japan to be a net buyer to completely solve the recycling issue because of the magnitude of FDI flows.

All that said, Japan's sitting on a big stock. And I think my interpretation of things that have happened is that as long as the short-term rate differential stays wide, a lot of Japanese institutions are reluctant to hedge. Some are willing to take a punt on a continued carry trade.

And then a whole set of institutions that I would think would be considering reallocating some of their foreign portfolio back to the US, I mean, back to Japan, given that the hedge yield ain't great. Well, maybe it'll get better if Stephen Myron gets his way and US rates come down to two next year. And the risk of holding an unhedged foreign bond portfolio seems very high, given that the real yen is roughly at its 1970s level.

You might think you would take advantage of higher nominal yields in Japan, a steeper curve, and bring money back. But no sign that that's happening. We've been surprised by that as well.

I guess with a new prime minister, it's that expectation is again pushed back because it looks like she wants to run a high-pressure economy. We will see if high pressure means that she's willing to accept yen 160 and more inflation. I mean, inflation has not been more popular in Japan than in the United States.

So I have some doubts. But for now, yeah, it looks like any catalyst for Japan will probably come from the, well, maybe a little bit from if she really is going to pull a Trump and say, I know monetary policy better than Ueda. If he may want to raid sites, but we're not going to let him.

But otherwise, it depends on the course of US rates. To me, the big issue is what happens when US rates really start coming down and hedging becomes cheap. I would think a lot of unhedged portfolios should hedge and should lock in rather massive profits.

So I wanted to discuss Europe, but we don't have time. And I feel a bit bad for skipping being Deutsche Bank, but we have to discuss the US. And so we discussed the big China being a big source of the global excess supply.

But I think anyone that looks at a simple chart, you have China as the accumulator of assets. The US is the big issuer. And Michael, you've been very critical of the role China's been playing as a big supplier of savings.

But what about the US? And fiscal policy, surely an independent decision. Actually, I would disagree with that. 6% plus deficits.

I'll get to that in a minute. I think when it comes to balance of payments, global balance of payments and global trade, I think it's one of the areas that we need to understand most. And it's also one of the areas in which there is probably no more nonsense than that particular area.

And Brad didn't pay me to say this, but I tell all my clients they got to read Brad's stuff because I think he's one of the few guys that really understands the global balance of payments. But the point that I would make in relationship to the United States is that as an American, I'm half American. I grew up in a lot of different countries, so maybe I have a slightly different attitude.

But I think Americans tend to exaggerate American agency in a lot of these things. So when Brad is talking, for example, about China, about Taiwan, Japan, it's very clear that their external imbalances are driven by domestic policy considerations and their control over their external imbalances. But in a world in which some countries control their external imbalances, in which case the internal imbalance always has to be consistent with the external imbalance.

So you develop your domestic policy and you control your external accounts so that you can implement that domestic policy. By definition, your external imbalance must be the external imbalance of your trade partners. And what I would argue there is that countries that exert more control over their external imbalances are able to initiate policies that then change the external imbalances of countries that exert less control.

Some of you may be thinking, oh, he's repeating Danny Roderick. Yeah, I'm repeating Danny Roderick. Danny Roderick said countries get to choose between more global integration and more economic sovereignty.

So if you look at Taiwan, let's say Taiwan and the US. So Taiwan has very high savings rate. How do I know?

Because it's running a current account surplus of 15% of GDP. So savings must be 15% percentage points higher than investment. And Taiwan decides where to invest that excess capital, that excess savings.

And for the most part, they invested in the United States. Now, notice what that means. The US doesn't control its capital account, right?

So it was a Taiwanese decision. It was a Beijing decision. It was a Seoul decision, a Berlin decision to invest excess savings in the US.

Technically, US, England, and Canada absorb something like 75% of all excess savings, very similar markets. But that has a really important implication. And the implication is that it is these surplus countries that are determining their external imbalances and therefore determining the external imbalances of the rest of the world.

And because the rest of the world's internal imbalances must be consistent with their external imbalances, then what's really driving the direction of causality is from countries that, to use Danny Roderick's phrasing, countries that exert more economic sovereignty towards countries that have chosen more along the lines of global integration. If you believe that, and I do, then the US current account deficit is not necessarily a function of domestic policies. So think about it this way.

You invest $100. Out of the blue, you invest $100 in the United States. By definition, the US economy must adjust.

There must be an additional $100 gap between investment and savings. One way the US could adjust is with $100 more investment, right? That's what happens typically developing countries.

That's what happened to the US in the 19th century. But the US doesn't have a saving constraint, right? American companies are sitting on piles of cash and they're not investing it.

They're buying back stocks. So if you give an additional $100 of investment, of saving, they're not going to say, oh, great. Now we can finally invest.

Investment won't go up. So somehow saving must go down, right? US has to adjust.

Now what Keynes told us and what Joan Robinson, and I think she's probably the best when it comes to trade, told us way back is that what ends up happening is that the country that invested $100, typically they got that through an expansion in manufacturing. They're able to run a surplus because their manufacturing exports are cheaper. So Americans buy $100 less American manufacturing.

American factory closes. Unemployment goes up. And you know, unemployed workers have negative savings.

So everything balances. Great. In the modern world, she wrote during the gold standard period when there were credit constraints.

But since the breakup of Bretton Woods in the early 70s, there was an alternative. And the alternative is if you don't want unemployment to go up, you can either accommodate an expansion in household borrowing, right? So borrowing goes up by $100.

Borrowing is negative savings. Or a fiscal expansion by $100. Again, negative savings.

So the US balances. If a foreigner decides to put $100 in the US, the US must decide to either increase investment by $100, increase unemployment, increase household debt, or increase the fiscal deficit, right? And I'm going to come to the fiscal deficit.

It's very important. Let's say the United States is able to get its house in order and reduce the fiscal deficit. What we are saying is that it's going to reduce the trade deficit, which also means it's going to reduce the net inflow of foreign capital into the United States.

And so for me, the question is, if the US exerts discipline on the fiscal deficit, will that really cause foreigners to be less interested in parking their savings in the US or more interested? Because if they are more interested because the US cuts its fiscal deficit, then by definition, the trade deficit will have to expand through some other mechanism, right? So the argument that the US fiscal deficit determines the trade deficit, that's not at all implicit in the accounting identity.

There's nothing that says that must be the case. What we know is that in some countries that can be the case and in other countries that cannot be the case, right? Otherwise, the world balances by sheer coincidence, which is very unlikely.

Does that make sense? I think Congress would be very happy with that description of the source of fiscal deficits. But irrespective of the question around agency, we saw in April that there's a tremendous interdependency and one thing we feel strongly in the research side here at DB is when thinking about debt sustainability, funding risk is a lot more about external balances and external funding rather than domestic saving rate.

You can see that in Japan, huge fiscal deficits, but the level of yield is very low because of the huge excess savings. So the question to you, Brad, is this fiscal concern in the US is obviously in the background, 6% fiscal deficit, potentially even wider if, for example, after the government shutdown, the debate is around more spending. If you take an external balance perspective, again, the net international investment position in the US keeps growing, that's accounting, keep running deficits.

Who's the marginal buyer of the treasuries? Because we talked about Taiwan, we talked about Japan. It's not a stable equilibrium in some sense.

So how do you think about that, the funding of the deficits? Both. Well, first, let me thank Deutsche Bank for remaining true to the balance of payments faith.

At a time when the IMF, you might think the Catholic Church of balance of payments analysis has lost the faith and no longer bothers to produce good balance of payments tables in emerging economies or for the world's second largest economy, China. I challenge any of you to read the IMF's last two staff reports on China and find any analysis of China's balance of payments. Rant over.

But thank you. I would nominate Deutsche Bank to be the next first deputy managing director of the IMF if we did a big swap. The interesting thing right now is that the current account deficit is not being funded primarily by people buying treasuries.

Treasury inflows have been a fraction of the current account for a reasonably long period of time. So it's probably inaccurate to think of a linear relationship between external financing and the treasury. What instead you're seeing, in part because of the hedging constraints, the cost of hedging, the flattish curve, is that more and more of the fixed income flow is coming from corporates and into corporate credit.

So it's not just the treasury market. And there is a little bit of a problem there implicitly because that implies that of the 6% of GDP fiscal deficit, now 5% of GDP of that is funded out of notes. 1% is covered by bills. Bills are mostly domestic.

Of that, at best, you're getting one and a half to two percentage points from the rest of the world. The other puzzle within this is that the countries with the big surpluses are not generally the countries that you see in the U.S. data buying the treasuries. So the last 12 months or so, you're going to see sales from China.

You're not seeing big inflows from Japan. You got a little bit from Taiwan, a little bit maybe from Korea. But the bulk of the flow is coming from London, is coming from Paris.

It is coming from the European custodial system. So in order to understand the flow, you have to be able in some sense, they don't have big surpluses, to link surplus, financial intermediary, deficit. And so that is my interpretation of all this, as well as the marginal buyer.

I think the marginal buyer, and I think this is consistent with a lot of like the boring IMF financial stability report kind of one. And it may not be a threat to financial stability, but I think there's more and more evidence that we have a long chain of risk intermediation. One set of players call it Chinese exporters hoarding dollars, is hoarding dollars short term.

And another set of players call it London hedge funds specializing in arbitrages involving derivatives relative to cash bonds, is the marginal actual buyer of treasuries. And the repo market allows people to run big cash bond balance sheets hedged. So that has been the trend.

And so far, it's been stable. But I think there's a lot of, I mean, if someone in this room has a better theory of the case, I'm actually all ears. I'd love to better understand this combination of flows.

But it has been stable, but it makes people nervous. And on top of that, you're seeing a big rotation in the external financing towards equity. Yeah, that's that story.

You guys have correctly pointed out there's some data questions about the magnitude of it. But in Q2 and Q3, our external deficit is funded by people buying into the AI bubble, period, which I do think matters for correlations going forward. So we don't have much time left.

I'd like to round off the conversation around US external balances with a discussion on tariffs. We started Liberation Day tariffs, came down, going back up. But speak to us, Brad, a bit about where do you understand the equilibrium?

Are the tariffs having an effect on the external balance? Are they both in terms of US levels, but then the distribution of those balances? China's 50% tariff still is even higher.

So are they material or is it not really having the impact we thought? Well, you do see a reallocation of the bilateral imbalance towards Southeast Asia, which has also gotten hit by pretty big tariffs. But that is yet to...

Would you call that transshipment? I do not call it transshipment. Transshipment is technically illegal.

It may be a little bit of it is illegal. I think it is parts going to final assembly. Technically, that's totally kosher.

I assume eventually rules will be written to kind of figure out how to tariff that. But for now, you can get your tariff rate down to 20. So what I think has happened, if you look at the details of the data, the biggest swings this year have come from pharmaceutical trade, which now is going to be excluded from the tariffs.

But that has led to enormous front running of imports in Q1 and is mostly responsible for current weakness, along with gold. So you got to look through all that. You look through all that, you will see a tariff impact on autos, wherein the auto guys are now seeking a little bit of tariff relief.

And you will see that being offset by imports related to AI, computers, chips, et cetera, which for now are still under a full exclusion from the tariffs. So on one level, you're not seeing adjustments. And I think it's unlikely you'll see adjustment if the AI bubble continues.

You can do an S&I or you can just look at the exclusions. And then I think the other relevant thing is, are we going to rebate the tariff revenue? If we rebate the tariff revenue, or whether to specific sectors that are adversely impacted, like autos, who US guys are hurt by the high steel tariffs, or to the broader population, which is starting to feel some places a little squeezed, there's no reason to think there would be much external adjustment.

Is this a stable equilibrium? There is never a stable equilibrium with tariffs, as long as Donald Trump is president. He will always want to have a tariff threat that he is using to put pressure on a US industrial sector, put pressure on a foreign trading partner.

On the other hand, have we reached a point in the cycle where the number of tariffs that are being reduced or a number of exclusions that are being granted may be close to equal to the number of tariffs that are being expanded? I think so. So your best guess of the underlying current account deficit, and we came into this around 4%, where are we coming out in a few months?

Around a little bit higher, around 4% is what I'm tracking. There's this craziness around the pharmaceutical sector, which is driving a little bit of that. We also are slowly seeing our income balance deteriorate.

And we are not yet at a rate constellation. Maybe we'll be next year with the rate cuts that Trump and Myron want that allows us to refinance at a lower cost our maturing bonds. So the 50% debt to GDP is just being refinanced every little.

So that's a mechanical upward push. So it turns out not much impact after all, the tariffs. Offsetting impacts at best.

I will go back to a question. I wanted to ask, but I was afraid I'd run out of time. We touched on the issue of technology and the AI flows, but Michael, you live in Beijing.

You've been there for 25 years. I think we all read of the American tech advantage on the chips and Nvidia, but talk to us about the US-China tech gap on the ground. And there's a narrative that China is actually ahead.

I'm not really a tech guy. I don't know what it needs to be ahead or not. What I think, though, is that for technology to be economically viable, it's got to be commercially viable.

And there's a real question about the commercial viability of technological investment in China. Now, there are a lot of people who think that China is the first country that has said we're going to rebalance through a massive productivity increase through investment in technology. And it's not true, right?

The Soviet Union tried the same thing in the 1960s. The Japanese were supposed to do the same thing in the 1990s. I don't know what the answer is, but I suspect the idea that technological advances make you rich may be a simplification.

Commercially sustainable technological advances may make you rich. What we're seeing in China for all of the investment in technology, worker productivity is going down pretty dramatically, and it's probably overstated. So maybe it'll make a difference.

Maybe not. I don't really know. But I do think if you keep spending $100 to produce $80 worth of value, you may create value for the world, but it's not clear that you're creating value for yourself.

That's interesting. We have a couple of minutes for questions. So I don't know if there's a mic going around or if everyone just wanted to raise their hand.

There is a mic. Is that a question at the back there? Yeah, there's a question there.

So there's talk of digital yuan and stable coins coming and obviously no digital dollar, but how does that affect if there was a digital dollar? How does that affect? You know, what we're talking about.

I don't, I'm sorry, I didn't understand the question. So if there was a digital dollar, how does that change things? Especially the direction of value of dollar.

You know, as the frictional cost of transferring currency is so low and a digital dollar would lower even more. I don't think that that's necessarily a good thing. I'm one of these guys who favor a kind of Tobin tax, but I often get that question about digital currency in China.

I often hear people say that China's got the secret plan to create a digital renminbi. And once it does, the renminbi will become the world's dominant currency. And the reason the renminbi is not a dominant currency is not because people don't know how to transact renminbi at low frictional costs.

The reason is because of capital controls. And I see no evidence at all that China wants to give up. It means basically giving up control of its external account and it clearly does not want to do that.

So I'm not sure digital currencies will have the big impact many people think it will. I don't know if, Brad, you have a view on digital currencies. I think a digital dollar is a dollar that doesn't pay interest.

And I am confused why so many people are so excited by an asset that doesn't pay interest. I'm being told there's workarounds by a financial engineer that allow that, which would be an interesting topic of further debate. With that, our time is up.

I believe there is lunch outside. And thank you very much, everyone. Podcept, the podcast from Deutsche Bank Research.

This podcast has been produced by Deutsche Bank and may contain research as defined in Method 2. The information discussed is believed to be reliable and has been obtained from public sources believed to be reliable, although Deutsche Bank makes no representation as to its accuracy or completeness. Opinions, estimates and projections discussed constitute the current judgment of the speaker at the time of recording.

They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. For further important information, please visit research.db.com.